Skip to content

AV-EXP-002 cross-ecosystem shadow calibration - #3

Merged
sneakocom merged 9 commits into
mainfrom
av-exp-002-shadow-benchmark
Aug 31, 2026
Merged

sneakocom merged 9 commits into
mainfrom
av-exp-002-shadow-benchmark

Conversation

@sneakocom

Copy link
Copy Markdown
Contributor

Summary

  • preregisters and executes the pinned Click / pytest-testmon shadow calibration
  • records 11-scenario FULL, ecosystem-selector, AV_CORE, and AV_WITH_SELECTOR_EVIDENCE evidence
  • reports one preserved AV selection miss and keeps trust at OBSERVE/SHADOW
  • adds invalid-state coverage, deterministic result verification, and bounded reproduction

Evidence

  • preregistration: f8a183c
  • result: sha256:5b3f99bfbebd3a0d061651d66adfb5a6aaef899475c6267cb20e4040e6ed5768
  • evidence bundle: sha256:435d8e5356ed6868edfc1747523ff75d1b327389bf14cc2675e867e45f4de705
  • AV_CORE: 86/87 relevant checks, one scenario miss
  • AV_WITH_SELECTOR_EVIDENCE: 86/87 relevant checks, one scenario miss

Verification

  • npm run verify
  • node benchmark/av-exp-002/verify-results.mjs

FULL remained authoritative throughout. No trust or lifecycle promotion is requested.

@sneakocom
sneakocom merged commit 3ff4168 into main Aug 31, 2026
2 checks passed
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant